6.4 Quiz: Web Analytics (EN&NL) Flashcards
There are three main types of web analytics: web usage mining, web content mining, and web structure mining.
True
Web usage mining
discovering interesting patterns in how visitors use a web site
Web content mining
extracting useful information or knowledge from web page contents
Web structure mining
mining hyperlink structure of web
Pan-session metrics are metrics that you calculate based on the entire session, including all individual pages that are visited.
False;
Pan-session metrics are metrics that consider the entire user session, but they do not necessarily include all individual pages visited. These metrics focus on the user’s behavior and interactions throughout the entire session, providing a broader view of user engagement and activity.
In contrast, page-level metrics focus specifically on individual pages and their performance, without considering the user’s journey across multiple pages within a session. It’s essential to differentiate between pan-session metrics, which look at the overall session, and page-level metrics, which analyze specific pages independently.
The bounce rate of a website is equal to the sum of the bounce rates of the individual pages out of which the website consists.
False;
Bounce rate: RATIO (aka percentage, not sum) of visits where visitor left instantly
bounce rate of site: ratio of single page view visits (or bounces) over total visits
bounce rate of specific page: single page view visits of that page over number of visits where that page was entry page
Leave out redirected and/or error pages
Ultimately, the bounce rate of a website will always be 100%.
False;
The bounce rate of a website will not always be 100%. The bounce rate is calculated as the percentage of visitors who navigate away from the site after viewing only one page, out of the total number of sessions. A 100% bounce rate would mean that every visitor left the site after viewing just one page, but this is not a universal constant.
Many websites have bounce rates that are less than 100%. Visitors may explore multiple pages during a session, leading to a lower bounce rate. A lower bounce rate is generally considered better, as it suggests that visitors are engaging with the content and navigating through the site. However, a 0% bounce rate is also uncommon, as some visitors may enter a site, find what they need on the first page, and exit without exploring further, contributing to a non-zero bounce rate.
Given the visits below: the bounce rate of page B is greater than the page exit rate of page B.
*Visit 1: A, E, B
*Visit 2: B
*Visit 3: B, E
*Visit 4: E
*Visit 5: B
*Visit 6: A, B
*Visit 7: B
*Visit 8: E
Bounce rate for page B: 3/4
(where B is entry page) 4 (v: 2, 3, 5, 7)
In 3 instances visitor legft immediatly (???)
–> bounce rate B; 3/4
Page exit rate for page B: 5/6
Page B was visited in visits 1, 2, 3, 5, 6 and 7. In 5 of those cases, the visitor left after having seen page B so resulting in a page exit rate of 5/6.
Bounce rate of site: 5/8
In server logs, HTTP requests are logged server side. Each HTTP request becomes one unique observation in the server log that shows, among other things, which page was requested at what time.
True
Server log analysis is a first way of gathering data for web analytics. As the terms suggests, the data is captured at the server side. Every HTTP request for a resource produces an entry in a Web server log file, which you can think of as a huge text file. This web server log file can then be parsed and processed to provide the useful web analytics information.
Page views are a more appropriate metric for marketing purposes than bounce rates or conversion rates.
False;
In many cases, a holistic approach that considers multiple metrics is beneficial. For example, while page views can showcase overall visibility, bounce rates can help assess the quality of engagement, and conversion rates provide insights into the effectiveness of the marketing funnel.
Ultimately, the choice of metrics should align with the specific goals and objectives of the marketing campaign, and a combination of metrics may be more informative than relying on any single metric alone.
In lead scoring or behavioral targeting, a website uses predictive analytics to show personalized content to users.
True;
By combining predictive analytics with lead scoring and behavioral targeting, websites can enhance the user experience by presenting content that is more likely to resonate with each visitor. This approach aims to increase engagement, conversions, and overall satisfaction by tailoring the online experience to the unique preferences and needs of individual users.
Time on site measurement is easier when using page tagging instead of server logs.
True
With off-site web analytics, your analyst has to analyze server logs from other pages in order to, for instance, estimate the visibility (share of voice) of your own website.
False;
Off-site web analytics typically involves analyzing data that is not directly collected from your own website but rather from external sources. This can include data from search engines, social media platforms, or other websites. In off-site analytics, server logs from other pages are not typically used.
To estimate the visibility or share of voice of your own website, analysts might rely on data from external sources such as search engine rankings, social media mentions, backlink analysis, or other third-party tools. These external data sources provide insights into how your website is perceived and how it performs in comparison to competitors on the broader internet landscape.
Server logs, on the other hand, are usually associated with on-site analytics, where data is collected directly from your own website’s server to analyze user behavior, interactions, and performance within your site.
Web analytics is one of the simplest application domains of analytics because the data is always available in a structured and cleaned manner.
False;
web analytics can be a complex field due to the diverse nature of data, the need for real-time analysis, and the challenges associated with ensuring data quality and privacy. While there are user-friendly tools available for basic analytics, handling more advanced analytics tasks often requires a good understanding of analytics principles and techniques.
When you as an analyst have access to server logs, you can typically perform more accurate analyses than when the website uses page tagging.
False;
When compared to web server log analysis, page tagging offers various advantages. First, it breaks through proxy servers and browser caching. In case of web server log analysis, if a page is cached, no record is logged on the web server which creates inaccuracies in the data collection. Page tagging allows to track client side events which is not supported by web server log analysis. Page tagging supports easy client-side collection of additional data, for example, by defining custom tags on an order confirmation page. This is not that easy for web server log analysis since integration with another database is required. Page tagging facilitates real-time data collection and processing whereas log files are typically analyzed in batch. Page tagging is often supported through hosted services which offers potential costs advantages. Furthermore, when using page tagging, the data capture is separated from the web design or programming. In other words, the JavaScript code for data collection can largely be updated by in-house analysts or the analytics service provider without the IT department having to implement changes. In case of web server log analysis, there is a larger reliance on the IT department to implement changes to capture more data. And finally, there are more innovation efforts put in by Web analytics vendors which is not the case for web server log analysis.
A Sankey diagram can be used to perform navigation analysis of users.
True
A Sankey diagram can be a useful tool for performing navigation analysis of users on a website or within an application. Sankey diagrams visually represent the flow of users or resources between different stages or pages, making it easier to understand the paths users take and identify potential points of interest or drop-off.
In the context of web analytics, a Sankey diagram may be employed to visualize user journeys, showing how users move from one page to another, and where they may exit the site. Each flow or connection between pages is represented by a link in the diagram, and the width of the links is proportional to the volume of users or interactions.
Given the visits below: the bounce rate and the page exit rate of page E is equal to 100%.
(Gegeven onderstaande visits: de bounce rate en de page exit rate van pagina E is gelijk aan 100%.)
Visit 1: A, E, B
Visit 2: B
Visit 3: B, E
Visit 4: E
Visit 5: B
Visit 6: A, B
Visit 7: B
Visit 8: E
False
The code below is an example of a page tag:
GET /hello.htm HTTP/1.1 User-Agent: Mozilla/4.0 (compatible; MSIE5.01; Windows NT) Host: www.tutorialspoint.com Accept-Language: en-us Accept-Encoding: gzip, deflate Connection: Keep-Alive
False
appears to be a snippet of an HTTP request message. Page tagging typically involves adding JavaScript code to web pages to track user interactions, collect data, and send it to analytics tools.
The code you’ve posted looks more like a part of an HTTP request header, where a client (e.g., a web browser) is requesting the page “/hello.htm” from the server. It includes information such as the User-Agent, Host, Accept-Language, Accept-Encoding, and Connection headers.
An example of a simple page tag might look like JavaScript code embedded in a web page, and it could be used for tracking user interactions. Here’s a hypothetical example:
html
Copy code
// Example page tag
var pageTagData = {
page: "/hello.htm",
userAgent: navigator.userAgent,
language: navigator.language,
// ... other data you want to track
};
// Send data to analytics tool (hypothetical function)
sendPageTagData(pageTagData);
</script>
